Purpose: Patient-reported pain and physical function are the most common outcomes examined in total joint replacement (TJR) for hip and knee osteoarthritis (OA). While TJR is an effective treatment for relieving pain and improving function, previous work has shown that the reporting of multiple symptomatic joints at the time of surgery negatively impacts 1-yr outcome scores. However, whether these negative effects persist beyond the recovery period is unknown. This work evaluated whether the presence of symptomatic joints predicted worse pain and physical function scores over a five year period following TJR. Methods: Participants completed a health questionnaire including the hip/knee-specific WOMAC pain and physical function measures pre-surgery and at 3, 6, 12 and 60 months post-surgery. Higher scores reflected worse status. Symptomatic joints were indicated on a homunculus pre-surgery based on pain or discomfort on most days of the previous month. Not including the surgical joint, symptomatic joint count was categorized as 0-1, 2-4, and 5+ (each group representing about one-third of the sample). Individuals reporting other hip or knee TJR during the 5 years of follow-up were not included in this analysis. Linear GEE regression modeling (accounting for the within-person correlated scores over time) was used to examine the relationship between symptomatic joint count and hip/knee-specific pain and physical function scores over the five years. Model covariates included age, sex, body mass index, comorbidity, hip/knee, and time since surgery. Results: 433 participants provided data up to and including the 5th year post-surgery. Age at surgery ranged from 35-88 years (mean = 65) with 61% female. At baseline 36% reported 0-1 symptomatic joints, 34% reported 2-4, and 30% reported 5+ symptomatic joints. No baseline differences in age, sex, surgical joint, BMI, or symptomatic joint counts were found between those who did and did not complete the 5th year survey. While there were statistical differences for baseline pain and physical function, the magnitude of these differences were small: 0.7 and 3 units on scales ranging from 0-20 and 0-68, respectively. Significant time trends for improvement in pain and physical function scores were found. Adjusting for covariates, increasing joint count was associated with worse pain and function scores throughout the 5 year follow-up (all p-values < 0.006), and more so for physical function. Accounting for baseline covariate effects, the least square mean differences at 5 years between those in the highest and lowest symptomatic joint count categories were nearly 2 for pain and 7.4 for physical function. Conclusions: While TJR is an effective intervention, these study results suggest that joint-specific patient-reported outcomes are negatively influenced by more than the symptoms of the surgical joint alone. Further, it appears that these negative effects are not limited to the early recovery period, and rather persist for many years following surgery. This knowledge should inform rehabilitation strategies post-TJR and the need for ongoing follow-up. The findings also potentially speak to a need for a more holistic approach to OA management.
Read full abstract